Variation and Life Strategies of the Trichoptera (Insecta) Larvae Community in a First Order Tributary of the Paquequer River, Southeastern Brazil

LIFE STRATEGIES OF TRICHOPTERA LARVAE 73

VARIATION AND LIFE STRATEGIES OF THE TRICHOPTERA (INSECTA) LARVAE COMMUNITY IN A FIRST ORDER TRIBUTARY OF THE PAQUEQUER RIVER, SOUTHEASTERN BRAZIL

HUAMANTINCO, A. A. and NESSIMIAN, J. L. Departamento de Zoologia, Instituto de Biologia, Universidade Federal do Rio de Janeiro, C.P. 68044, CEP 21944-970, Rio de Janeiro, RJ, Brazil Correspondence to: Jorge Luiz Nessimian, Departamento de Zoologia, Instituto de Biologia, Universidade Federal do Rio de Janeiro, C.P. 68044, CEP 21944-970, Rio de Janeiro, RJ, Brazil Received February 23, 1999 – Accepted May 24, 1999 – Distributed February 28, 2000 (With 5 figures)

ABSTRACT The Trichoptera fauna was sampled in a first order tributary of Paquequer River, in the district of Teresópolis, Rio de Janeiro State. Quantitative samples of litter from pools, litter from riffles, sand, and stones were taken on each season from February 1991 to February 1992.The life cycles of common genera were categorized into three groups of biological strategies. The largest group represented the general profile of the community, in which ultimate instar larvae were predominant in spring and the younger ones in summer. Probably, emergence and oviposition occur in spring. Key words: Trichoptera, larvae, life strategies, streams, Atlantic Forest.

RESUMO Variação e estratégias de vida da comunidade de larvas de Trichoptera (Insecta) em um tributário de primeira ordem do Rio Paquequer, Sudeste do Brasil Na seção de primeira ordem do Rio Paquequer, Teresópolis, RJ, foram tomadas amostras de 4 tipos de substrato: litter depositado em zonas de remanso, litter retido em zonas de correnteza, pedra e areia. De cada tipo foram coligidas 5 amostras em cada estação do ano, de fevereiro de 1991 a fevereiro de 1992. Através de dados de freqüência e biomassa de cada gênero, alguns aspectos da história de vida foram inferidos. Com base nesses dados, foram separados três grupos estratégicos. O maior deles reflete o perfil geral da comunidade, no qual os últimos estádios larvais foram predominantes na primavera e os estádios iniciais, no verão. Provavelmente, a emergência e a oviposição ocorrem na primavera. Palavras-chave: Trichoptera, larvas, estratégias de vida, riachos, Mata Atlântica.

INTRODUCTION emergence is under the dual control of temperature and length of the day, and presumably the relative The knowledge of life cycles of the aquatic importance of these factors varies from place to insects is essential for understanding most of the place. ecological features of this fauna. Wiggins (1996) When compared with temperate areas, tropical pointed out that most of the Trichoptera in habitats present smaller variations in their climatic temperate latitudes complete a generation every factors, and they are generally considered less year. According to Hynes (1970), the time of seasonal to many organisms (MacArthur, 1972

Rev. Brasil. Biol., 60(1): 73-82 74 HUAMANTINCO, A. A. and NESSIMIAN, J. L.

apud: McElravy et al., 1982). According to Yule & Smicridea spp. the distribution of larval sizes Pearson (1996), there are few seasonality studies indicated multivoltine life cycles. However, the on Trichoptera carried out in tropical streams with seasonal abundances varied, showing that the a humid-dry regime. These studies are based only regime of precipitations, and in a smaller degree the on data of adult emergence (McElravy et al., 1982 temperature, caused a certain seasonality. and Benson & Pearson, 1988; Dudgeon, 1988; The goal of this work is (1) to analyze the Statzner, 1976; apud: Yule & Pearson, 1996) and populational variations of the most abundant report similar patterns which are mostly, seasonal. Trichoptera genera in a section of first order of a In the neotropics, Wolf et al. (1988), in a study river in the montane Atlantic Forest area of concerning the emergence of aquatic insects in southeastern Brazil and (2) the relation between Colombia, found that the largest emergence came, environmental factors and life cycles. generally, after the high pluviosity periods and they observed that temperature and rain seemed to be STUDY AREA the main regulatory factors. Flint & Masteller (1993), working in a stream in Puerto Rico, found that The sampling was carried out in a first order distinct correlations did not exist between the tributary of the Paquequer River (ca. 1,100 m high), pattern of pluviosity and the emergence of Teresópolis (22°24’44”S and 42º57’56”W), State of Trichoptera. Although they observed that some Rio de Janeiro (Fig. 1). The abundant riparian seasonality seemed to occur in the flight periods vegetation maintains most of the stream under of some species, the most abundant ones were shade. The stream width ranges from about 0.5 m present throughout the year. In Brazil, Oliveira & to 2.5 m and its bed is constituted mainly by sand, Froehlich (1997), studying the fauna of Trichoptera pebbles and rocks. The allochthonous organic in a stream of the cerrado region in the State of São matter (litter), is retained in riffles or deposited in Paulo, found that for Leptonema spp. and pools.

ESPÍRITO SANTO 21 N MINAS GERAIS do Rio Paraíba Sul

Sul 22 do

Paraíba RIO DE JANEIRO Rio Rio Paquequer « OCEANO ATLÂNTICO SÃO PAULO

80 km Rio de Janeiro 23

44 43 42 41

Fig. 1 — Map of the State of Rio de Janeiro showing the location of the area of studies (Rio Paquequer, Teresópolis).

Rev. Brasil. Biol., 60(1): 73-82 LIFE STRATEGIES OF TRICHOPTERA LARVAE 75

MATERIAL AND METHODS of the water were measured with a portable conductivimeter (Analion). The time a floating object Twenty samples were taken for the quan- took to cover a fixed distance in a stream section titative sampling in each season from February 1991 with known depth and width was used to measure to February 1992. The analyzed months were the discharge. February, April, July, and October 1991 and In order to verify the relations among the February 1992, totalizing 100 samples. The most abundant genera, in regard to their life cycles sampling area of each collection was approximately and strategies, the covariance was quantified 1.25 m2. Four substrate types were considered: among all pairs of values of a matrix that contained sand, litter retained in riffles, litter deposited in the frequency and biomass percentages. The pools, and stones. Five samples were taken from Pearson correlation coefficient was used (Ludwig each substrate, using a Surber dredge (0.3 mm & Reynolds, 1988). The values of the obtained mesh) for stones and a D-net (1 mm mesh) for the covariance were used in Cluster Analysis, by the other substrata, both with 0.05 m2 of collecting UPGMA method (Sneath & Sokal, 1973). area. The collected material was fixed in 4% formaldehyde and conserved in 80% ethanol. RESULTS The collected material was fixed, separated and counted under a stereomicroscope (160x max.). Environmental variables The species or morphospecies were identified with The values of the physical and physical- the keys and diagnoses of Angrisano (1995), chemical variables considered in the present study Wiggins (1996), and the works of Flint (1969), Flint showed that the atmospheric temperatures were & Bueno-Soria (1982), Flint & Wallace (1980), similar in the summer and in the autumn of 1991. Guahyba (1981), and Holzenthal (1988a,b). The lowest temperatures, as expected, occurred in Comparisons with deposited specimens in the the winter. The summer of 1992 presented lower Museu Nacional (UFRJ, Rio de Janeiro) were made temperatures than those of the previous year. The to confirm some identifications. Part of the material water temperature accompanied the atmospheric was identified with the help of prof. R. Holzenthal, one in its general profile. (University of Minnesota) and prof. R. Guahyba The pH values were lightly acid and with little (Museu Nacional, UFRJ). variation. The largest value occurred in the autumn Early instar larvae that did not allow a reliable of 1991 (5.8), and the smallest in the summer of identification were grouped in the family category. 1992 (4.9). The dissolved oxygen concentrations The larvae with very low frequencies were not presented the largest value in autumn (11.25 considered in the statistical analysis. mg.l–1) decreasing until February of 1992 (5.7 The biomass of the larvae was estimated mg.l–1). The electric conductivity varied a little during using the humid weight of the fixed material. The the period of study. It presented a larger value in individuals of each species or morphospecies were the spring of 1991 (21.3 (mS.cm–1) that is explained separated into size classes. For each size class of by the largest amount of material carried during the each species a number of individuals (10 in first rains after the dry period (winter). The average) was weighed in an analytic lever (0.1 mg discharge varied between 0.039 m3.s–1 (autumn of of accuracy) to obtain the medium weight value 1991) and 0.006 m3.s–1 (winter of 1991), reflecting (Nessimian, 1995). The larvae were weighed without the pluviosity and separating two different periods their cases. (rainy and dry). At each collection, temperature, pH, dissolved oxygen, electric conductivity, and The community composition and the populational discharge were taken. A mercury thermometer was variation of the most abundant groups used to measure the water temperature. The pH A total of 2.582 larvae (total weight of 10.4 g). values were obtained with the aid of a portable pH- belonging to 9 families and 20 genera were meter (Analion). The values of dissolved oxygen analyzed. The families Leptoceridae (60.14%), with were measured by the method of Winkler (Brower the genera Nectopsyche (35.36%), and Triplectides & Zar, 1977). The values of electric conductivity (16.65%), and Hydropsychidae (22.45%), with the

Rev. Brasil. Biol., 60(1): 73-82 76 HUAMANTINCO, A. A. and NESSIMIAN, J. L.

genus Smicridea (18.90%) were the most abun- big was found, in autumn and winter, while in dant. spring only an individual of medium size was Phylloicus (Calamoceratidae). The larvae of collected. In the summer a very reduced number of this genus occurred during the whole year. The small larvae was collected (Fig. 2). frequency and biomass peaks appeared in winter, Smicridea (Hydropsychidae). This was the corresponding to the months of drought. Larger most important genus in substrata submitted to larvae were observed in spring, coinciding with the high-speed currents (retained litter and stones). increase in discharge and possibly with adult The largest frequency values were obtained in emergence and oviposition. In the following autumn. The same happened with the biomass season, as expected, the smaller larvae were the values. The largest larvae prevailed in the summer predominant (Fig. 2). of 1991 and the smallest ones, in autumn. It is Helicopsyche (Helicopsychidae). The larvae probable that oviposition happened in the begin- of this genus, which occur typically in areas with ning of the rains (Fig. 2). high current, were collected in the stony substratum Atanatolica (Leptoceridae). The presence of and retained litter. Larger frequency and biomass these larvae was closely related to the summer. In values were observed in the summer of 1992. The the summer of 1991 small larvae prevailed. In the smaller larvae occurred in larger numbers in the following seasons, they did not appear in the autumn, coinciding with the largest discharge. In samples. In the summer of 1992 grown larvae the following months the growth of the biomass occurred. The cycle can be considered slow was verified up to spring, when the developed (Fig. 2). larvae prevailed and probably adult emergence and Nectopsyche (Leptoceridae). This very abun- oviposition happened. The life cycle seems to be dant genus constituted 1/3 of the collected larvae. slow and univoltine (Fig. 2). They presented a frequency peak in winter and a Atopsyche (Hydrobiosidae). The larvae biomass peak in spring. The small larvae pre- occurred mainly in the autumn, the season of larger dominated in winter and the large ones in spring, discharge, and were absent in the winter samples. indicating summer and autumn as the probable This genus seems to present a short cycle, with season of adult emergence (Fig. 2). more developed larvae and a biomass peak in the Notalina (?) (Leptoceridae). In this genus summer of 1991 and a prevalence of young larvae there was a predominance of larger larvae for occurring in the following season, when the almost the whole study period, except in the summer frequency peak occurred. Adult emergence and of 1992, when the smallest ones prevailed. The oviposition possibly took place before the largest peak, both in frequency and biomass, beginning of the abundant rains of the autumn of happened in winter. Probably, as in some previous 1991 (Fig. 2). groups, the oviposition are related to the first rains Blepharopus (Hydropsychidae). The largest in the spring (Fig. 3). frequency of the larvae coincided with the season Oecetis (Leptoceridae). As in other Lepto- of larger pluviosity. The population seems to show ceridae, these larvae were strongly associated to an annual cycle. In the summer the less developed the litter retained in riffle areas. In summer and larvae prevailed, while in the autumn and in the autumn the population was constituted for the winter the larger ones were more numerous. The most part by small larvae. In winter, the frequency oviposition seems to coincide with the spring and biomass peaks coincided. In the following rains, since the frequency of small larvae increases season, the larvae reached their largest sizes and quickly from spring to summer (Fig. 2). the adult emergence and oviposition probably Leptonema (Hydropsychidae). This is one of occurred (Fig. 3). the genera with a larger contribution to the Triplectides (Leptoceridae). The larvae of this community biomass. The graphs of frequency and genus, as the above mentioned Nectopsyche, were biomass percentages of the larvae of Leptonema the most abundant in the deposited litter. An do not show, as in most of the other genera, a clear analysis of the populacional fluctuation graphs prevalence of small or big larvae in a given season. suggests that they apparently avoid competition However, a larger amount of larvae both small and by means of temporal segregation.

Rev. Brasil. Biol., 60(1): 73-82 LIFE STRATEGIES OF TRICHOPTERA LARVAE 77

Phylloicus Helicopsyche

40 60 50 30 40 % 20 % 30

10 20 10 0 0 V91 O I P V92 V91 O I P V92

Atopsyche Blepharopus

80 50 70 40 60 50 30 % 40 % 30 20 20 10 10 0 0 V91OOIIPPV92 V91 V92

Leptonema Smicridea

60 50 50 40 40 30 % 30 20 20 10 10 0 0 V91OOIIPPV92 V91 V92

Atanatolica Nectopsyche

90 60 80 50 70 60 40 % 50 %30 40 30 20 20 10 10 0 0 V91 O I P V92 V91 O I P V92

Fig. 2 — Frequency (----) and biomass (——) percentages of Trichoptera larvae in a first order tributary of Paquequer River, Teresópolis, RJ, in the study period. V91 – summer of 1991; O – autumn; I – winter; P – spring; V92 – summer of 1992.

Rev. Brasil. Biol., 60(1): 73-82 78 HUAMANTINCO, A. A. and NESSIMIAN, J. L.

Notalina (?) Oecetis

30 35 25 30 25 20 20 %%15 15 10 10 5 5 0 0 V91 O I P V92 V91 O I P V92

Triplectides Wormaldia

60 70 50 60 40 50 40 % 30 % 30 20 20 10 10 0 0 V91 O I P V92 V91 O I P V92

Barypenthus

% 20

0 V91 O I P V92

Fig. 3 — Frequency (----) and biomass (——) percentages of Trichoptera larvae in a first order tributary of Paquequer River, Teresópolis, RJ, in the study period. V91 – summer of 1991; O – autumn; I – winter; P – spring; V92 – summer of 1992.

The largest values in frequency and biomass smaller larvae also prevailed. The latter happened occurred in the summer of 1992. Developed larvae in spring. The rains that occurred in spring prevailed in autumn and the more young ones in apparently influenced the adult emergence and the summer of 1992. Univoltinism probably occurs oviposition, a fact that can be noticed due to the (Fig. 3). prevalence of young larvae in the following summer Barypenthus (Odontoceridae). The larvae of (Fig. 3). this genus are of large size, contributing with more Wormaldia (Philopotamidae).These larvae than 50% of the community biomass. The were met only in substrata submitted to high frequency peaks and biomass did not coincide in currents. They are associated to autumn, the time. The former happened in winter, when the season of larger discharge.

Rev. Brasil. Biol., 60(1): 73-82 LIFE STRATEGIES OF TRICHOPTERA LARVAE 79

Small larvae were collected only in autumn, adult emergencies and oviposition of most of the coinciding with the frequency peak. Adult species occur, resulting in the predominance of emergence and oviposition appear to have young larvae in the following season (Fig. 4). occurred in the begining of the rains of the autumn The cluster analysis applied to the frequency of 1991, when the larger larvae prevailed (Fig. 3). and biomass percentages of 15 taxa resulted in the recognition of two main temporaly groups that Relation between life cycle and environment denote different life strategies: the winter-spring The whole Trichoptera community presented group and the summer-autumn group. an increase in the frequency of larvae from the The winter-spring group contains larvae that summer to the winter of 1991. The frequency then reach their largest sizes in spring and they follow started to decline, reaching its smallest value in the the general pattern of the community (Nectopsyche, spring. From the spring to the summer of 1992 an Phylloicus, Barypenthus, and others). The summer- increase in the frequency was verified, so that the autumn group is divided into two branches: the peak was achieved in the latter season. The biomass summer group, which is characterized by the small presented its smallest value in autumn, starting then number of individuals during autumn, winter, and to grow until the spring (the largest value found). spring, but presents peaks of frequency and/or As printed out above, during autumn and biomass in summer (Helicopsyche, Triplectides, and winter the community grows both in frequency and Atanatolica); and the autumn group, characterized biomass. In these seasons, young larvae are by a strong prevalence of small larvae in the autumn. predominant. In the spring the frequency curve goes The adult emergence and oviposition of eggs down abruptly, while that of biomass increases. In should have happened in the summer of 1991 this season larger larvae prevail and, probably, (Smicridea and Wormaldia) (Fig. 5).

% 15

0 V91 O I P V92

Fig. 4 — Frequency (----) and biomass (——) percentages of the Trichoptera larvae community in a first order tributary of Paquequer River, Teresópolis, RJ, in the study period. V91 – summer of 1991; O – autumn; I – winter; P – spring; V92 – summer of 1992.

DISCUSSION involving the microorganisms and aquatic invertebrates. Nolen & Pearson (1993), working in Environmental variables a stream of the tropical humid forest in Australia, As in other studies carried out in first order found that a shredder species of Calamoceratidae, streams of tropical zones, the values of the water processed leaves more quickly when temperatures temperature presented a small variation. Several were higher during summer. authors, as Wolf et al. (1988), found that the A variation was observed in the values of temperature does not seem to be a fundamental dissolved oxygen. This variation is directly related factor in the structure and distribution of aquatic to discharge, as the studied area is fundamentally organisms in the tropics. However, the temperature of high pluviosity. The values are close to can play na important role in several processes saturation.

Rev. Brasil. Biol., 60(1): 73-82 80 HUAMANTINCO, A. A. and NESSIMIAN, J. L.

Oe Le

Hy Sm

–0.500 –0.250 0.000 0.250 0.500 0.750 1.000

Fig. 5 — Dendrogram based on the values of Pearson correlation coeficient (r = 0.497, p = 0.05) calculated among the most frequent genera in the sampling during the study period. At – Atanatolica; Ap – Atopsyche; Ba – Barypenthus; Bl – Blepharopus; He – Helicopsyche; Hy – Hydropsychidae; Le – Leptonema; Lp – Leptoceridae; Nc – Nectopsyche; No – Notalina; Oe – Oecetis; Ph – Phylloicus; Sm – Smicridea; Tr – Triplectides; Wo – Wormaldia.

The pH values were lightly acid, with a temperature (although slight). The adult emer- medium value of 5.3. Although low pH values are gences that happen in this period would be related, in several cases, with smaller diversity of associated to a syncronism (Oliveira, 1996) as an species and lower productivity (Ward, 1992), studies adaptation to the environment, favouring, due to in New Zealand (Winterbourn & Collier, 1987 apud: the new rains, the oviposition and dissemination Ward, 1992) in streams of natural acidity showed of the eggs and larvae. that only values lower than 4.6 were able to reduce According to Oliveira (1996), who worked on the taxonomic richness to 50%. the Trichoptera community in streams of the das The discharge data are correlated directly with Almas river basin (Goiás) in the months subsequent the pluviosity which is high in summer and lower to the colder temperatures and before the beginning in winter. These features suggest an unimodal of rains (specifically the months of September and pluviosity regime. October), an evident increase in the amount of all instars occurred, suggesting the possibility of a Seasonality of the biological cycle “regulator period” in the distribution of those taxa. For most of species of Trichoptera, autumn According to Flint (1991a), who obtained data on and winter are seasons when the population growth the emergence of many taxa during one year in is constant with a prevalence of young larvae, Colombia, the emergence in Trichoptera is clearly which exploit the food carried by the current in correlated with the rains in the study area, where autumn, and the leaves deposited in pools, when the smallest pluviosities happen in January and the smaller pluviosity results in good quality there are peaks in April and in October (bimodal detritus. In the spring larger larvae prevail, pattern). The emergence also presented peaks from coinciding with the beginning of the rains after the April to May. Undoubtedly, some specific event, months of drought and the increase of the water which happens in the period February-March and

Rev. Brasil. Biol., 60(1): 73-82 LIFE STRATEGIES OF TRICHOPTERA LARVAE 81

is related with the rain, starts the emergence process FLINT, O. S., Jr. & MASTELLER, E. C., 1993, Emergence that attains a peak one or two months later. composition and phenology of Trichoptera from a tropical rainforest stream at El Verde, Puerto Rico. J. With regard to the strategic groups, the Kan. Ent. Soc., 66(2): 140-150. winter-spring group, which follows the community FLINT, O. S., Jr. & WALLACE, J. B., 1980, Studies of general profile, seems to be synchronized with the Neotropical Caddisflies, XXV: The immature stages of adult emergence in favorable climatic conditions Blepharopus diaphanus and Leptonema columbianum (spring), as well as with the temperature increase, (Trichoptera: Hydropsychidae). Proc. Biol. Soc. Wash., the beginning of the rains, and the input of organic 93(1): 178-193. matter in the stream. In this group shredder larvae GUAHYBA, R. R., 1981, Contribuição ao estudo dos (e.g., Nectopsyche and Phylloicus) and predators Trichoptera KYRBY, 1813, na represa dos ciganos, Rio de Janeiro, baseada em larvas, pupas e casas as (e.g., Barypenthus and Oecetis) prevail. The (INSECTA). Dissertação de Mestrado, MN-UFRJ, Rio larvae of the summer group would be synchronizing de Janeiro, 156p. their frequency peaks with the largest periphyton HOLZENTHAL, R. W., 1988a, Systematics of Neotropi- availability in this season. In this group we have cal Triplectides (Trichoptera: Leptoceridae). Ann. Ent. the scrapers Helicopsyche and Atanatolica. The Soc. Am., 81(2): 187-208. life cycles included in the autumn group seem to HOLZENTHAL, R. W., 1988b, Studies in Neotropical synchronize the new larvae with the largest Leptoceridae (Trichoptera), VIII: The Genera Atanatolica Mosely and Grumichella Muller (Tri- discharge observed in the study period. Smicri- plectidinae: Grumichellini). Trans. Am. Ent. Soc., dea, the collector of largest frequency in the 114: 71-128. community, is a member of this group. HYNES, H. B. N., 1970, The ecology of stream insects. Ann. Rev. Ent., 15: 25-42. Aknowledgments — The authors thank Prof. Dr. Leandro G. Oliveira (UFG), Prof. Dr. Alcimar L. Carvalho (Museu LUDWIG, J. A. & REYNOLDS, J. F., 1988, Statistical Nacional, UFRJ), Profa. Dra. Marcia S. Couri (Museu Ecology. J. Wiley & Sons, New York, xviii+337p. Nacional, UFRJ), and Prof. Gabriel Mejdalani (Museu McELRAVY, E. P., WOLDA, H. & RESH, V. H., 1982, Nacional, UFRJ) for useful comments and suggestions on Seasonality and annual variability of caddisflies adults the manuscript. This study was suported by FAPERJ, FUJB, (Trichoptera) in a “non-seasonal” tropical environ- and CNPq. ment. Arch. Hydrobiol., 94(3): 302-317. NESSIMIAN, J. L., 1995, Composição da fauna de inver- REFERENCES tebrados bentônicos em um brejo entre dunas no litoral do Estado do Rio de Janeiro, Brasil. Acta ANGRISANO, E. B., 1995, Insecta Trichoptera, pp. Limnol. Bras., 7: 41-59. 1199-1237 In: E. C. Lopretto & G. Tell (eds.), NOLEN, J. A. & PEARSON R, G., 1993, Factors affecting Ecosistemas de Aguas Continentales: metodologias litter processing by Anisocentropus kirramus (Tri- para su estudio, v. III, xvii+897-1401 p., Ediciones choptera: Calamoceratidae) from an Australian tropi- Sur, La Plata. cal rainforest stream. Freswat. Biol., 29: 469-479. BROWER, J. E. & ZAR, J. H., 1977, Field and Labora- OLIVEIRA, L. G., 1996, Aspectos da biologia de comu- tory Methods for General Ecology. Wm.C. Brown nidades de insetos aquáticos da ordem Trichoptera pub. Dubuque, 2a ed., 226p. Kirby, 1813, em córregos de cerrado do município de FLINT, O. S., Jr., 1969, Studies of Neotropical Pirenópolis, Estado de Goiás. Tese de Doutorado, Caddisflies, VIII: Immature Stages of Barypenthus Instituto de Biociências da Universidade de São Paulo, claudens. Proc. Ent. Soc. Wash., 171(1): 24-28. São Paulo, ii+120p. FLINT, O. S., Jr., 1991, Studies of Neotropical OLIVEIRA, L. G. & FROEHLICH, C. G., 1997, The Caddisflies, XLV: The Taxonomy, Phenology and Trichoptera (Insecta) fauna of a “cerrado” stream in faunistics of the Trichoptera of Antioquia, Colombia. southeastern Brazil. Naturalia, 22: 183-197. Smithson. Contr. Zool., 52: 1-113. SNEATH, P. H. A. & SOKAL, R. R., 1973, Numerical FLINT, O. S., Jr. & BUENO-SORIA, J., 1982, Studies of Taxonomy – The principles and practice of numerical Neotropical Caddisflies XXXII: The immature stages classification. W. H. Freeman & Co., San Francisco, of Macronema variipenne Flint & Bueno, with the xv+573p. division of Macronema by the resurrection of WARD, J. V., 1992, Aquatic Insect Ecology. 1- Biology Macrostemum (Trichoptera:Hydropsychidae). Proc. and Habitat. J. Wiley & Sons Inc., New York, Biol. Soc. Wash., 95(2): 358-370. xi+438p.

Rev. Brasil. Biol., 60(1): 73-82 82 HUAMANTINCO, A. A. and NESSIMIAN, J. L.

WIGGINS, G. B., 1996, Larvae of the North American YULE, C. M. & PEARSON R. G., 1996, Seasonality of Caddisfly Genera (Trichoptera). Univ. of Toronto benthic invertebrates in a tropical stream on Press, Toronto, 2a ed., xiii+457p. Bougainville Island, Papua New Guinea. Arch. Hydrobiol., 137(1): 95-117. WOLF, M., MATTHIAS, U. & ROLDAN, G., 1988, Estudio del desarrollo de los insectos acuaticos, su emergencia y ecologia en tres ecosistemas diferentes en el Departamento de Antioquia. Actualidades Biologicas, 17(63): 2-27.

Rev. Brasil. Biol., 60(1): 73-82